Date: Fri, 5 Jan 2018 20:29:45 +0300 From: Slawa Olhovchenkov <slw@zxy.spb.ru> To: Eugene Grosbein <eugen@grosbein.net> Cc: svn-src-head@freebsd.org, Steven Hartland <smh@FreeBSD.org>, src-committers@freebsd.org, svn-src-all@freebsd.org Subject: Re: svn commit: r327559 - in head: . sys/net Message-ID: <20180105172945.GC5368@zxy.spb.ru> In-Reply-To: <5A4FAEF2.8000802@grosbein.net> References: <201801042005.w04K5liB049411@repo.freebsd.org> <5A4E9397.9000308@grosbein.net> <20180105131105.GB5368@zxy.spb.ru> <5A4F7F70.5050106@grosbein.net> <20180105143825.GL19373@zxy.spb.ru> <5A4FAEF2.8000802@grosbein.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Jan 05, 2018 at 11:59:30PM +0700, Eugene Grosbein wrote: > 05.01.2018 21:38, Slawa Olhovchenkov wrote: > > >>> Irrelevant to RSS and etc. flowid distribution in lacp case work very > >>> bad. This is good and must be MFC (IMHO). > >> > >> It may work bad depending on NIC and/or traffic type. > >> It works just fine in common case of IP forwarding for packets with TCP/UDP inside. > >> > >> It can be easily disabled locally for specific cases when it does not work. > > > > Packet distrubuting on network equipment (lacp case) w/ enabled flowid cause > > uneven queue distributing. > > Once again: this heavily depends on local environment and this is not true for many cases. For any case resolving cavears need uncommon skills. It's true. Don't using flowid from hardware need negligible CPU power. Safe is disable flowid bDefault disable flowid is safe way w/o performace penalty.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20180105172945.GC5368>